The Need For AI Technologies in Elder Care

According to the US Census Bureau, the number of older adults is anticipated to double from 45 million (2019) to 90 million by 2050. This is a tremendous shift and one of the primary concerns of nursing home staff amid the nationwide healthcare staff shortage is health outcomes, particularly when it comes to falls and fall injuries.

Using AI technology to enhance the delivery of care has the potential to bring about significant transformations, both in terms of economic benefits and improved health outcomes. According to a joint report from McKinsey & Company and Harvard University, greater adoption of artificial intelligence in healthcare could save the U.S. healthcare system between 5% and 10% annually, which is equivalent to $200 billion to $360 billion in annual spending.

The graph below shows a review of 550,000 healthcare-related patents grouped into 24 categories with the innovation stage ranked as emerging, accelerating, and maturing. As one can see, AI-assisted EHR/EMR is at the forefront of the accelerating innovation stage and just behind the three maturing technologies. Rightly so as the use of AI technology can assist organizations in analyzing data rapidly and obtaining valuable insights that may have previously been unavailable.

Missing in the graph are fall detection and fall prevention technologies. As a quick exercise, I searched the United States Patent and Technology Office database for patents related to AI (or machine learning), fall detection, and prevention and only found 1400 patents. Clearly, with falls among seniors being the leading cause of accidental deaths among seniors in the United States with a medical price tag of $50 billion dollars each year, there is a tremendous gap in care technologies.

This blog discusses innovative AI-driven technologies that have entered the nursing home space aimed to lessen the workload and improve the daily routine of healthcare professionals, enabling them to focus more on fulfilling the healthcare requirements of their patients.

The Problem with Status-quo Technology Solutions

When a senior falls, every second counts. Every 11 seconds, a senior falls and is seriously injured requiring a trip to the emergency room. Fall detection technologies have often been used to get help to the senior as soon as possible. The reason is that seniors who receive help more than 1 hour after a fall have a greater risk of complications such as dehydration, pressure sores, pneumonia, hypothermia, and rhabdomyolysis. By the way, rhabdomyolysis is a serious medical condition that can be fatal or result in permanent disability according to the Centers for Disease Control and Prevention.

Traditionally, fall detection mats placed on beds, chairs, and floors were used to detect falls.

However, the detection mats had negative unintended consequences for patients as identified in the Centers for Medicaid and Medicare Services guidance document for surveyors of long-term care facilities (Rev. 211, 02-03-23).

“Examples of negative potential or actual outcomes which may result from the use of position change alarms as physical restraint, include:

Loss of dignity;
Decreased mobility;
Bowel and bladder incontinence;
Sleep disturbances due to the sound of the alarm or because the resident is afraid to move in bed thereby setting off the alarm;
Confusion, fear, agitation, anxiety, or irritation in response to the sound of the alarm as residents may mistake the alarm as a warning or as something they need to get away from.

We should keep in mind that there are special circumstances where fall detection mats and other devices are required to keep residents safe. According to the guidance document,

“a resident may have a device in place that the facility has stated can be removed by the resident.”

The Problem with Using Sitters

In healthcare settings, sitters are employed to continuously monitor patients who are at risk of harming themselves or others. They are also utilized for patients who may experience confusion or are at risk of falling or wandering. In nursing homes, there is often a budgetary challenge when it comes to hiring sitters. Therefore, nursing and care staff are often tasked with keeping a watchful eye on these at-risk patients. It is typical to find multiple seniors in wheelchairs near the nurse's station, while nurses manage and document patient health records behind the desk.

Using sitters to prevent falls is not a sustainable solution for several reasons:

Cost

Hiring sitters to monitor patients is expensive and can be financially burdensome for patients and their families, especially if they require around-the-clock care.

Limited availability

The nationwide staff shortage of healthcare workers can make it difficult to find sitters. Moreover, sitters may not be available to keep their eyes on the patient at all times, which could leave gaps in care and place the patient at risk of falling.

Reduced independence

Using sitters to prevent falls can limit patients' independence and autonomy, which can have a negative impact on their emotional well-being.

Ethical concerns

The use of sitters can raise ethical concerns, as patients' are 'watched' beyond their comfort level and invasion of privacy becomes an issue.The use of sitters to prevent falls is not a sustainable solution due to cost, limitation on patients' independence, raising ethical concerns, and may not always be effective in preventing falls.

According to a 2021 study by Dr. Adela Greeley (West Los Angeles Veterans Affairs Medical Center and David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, California)

Detecting Falls with AI-Based Wrist-Worn Wearables

Long gone are the days when a senior has to wear a stigmatizing “Help, I’ve fallen and I can’t get up.” fall detection pendant.

Now, AI technology can be used to detect falls in nursing homes by using sensors placed throughout the facility. These sensors can detect changes in movement patterns to predict future falls or alert staff when a resident does fall.

For example, wrist-worn devices can detect movements like walking, running, and standing. The data collected by the band is transmitted to a beacon hub which then sends the information to secure cloud servers.

Their AI analyzes data to detect changes in habits related to activities of daily living so that nurses can use it to detect falls as well as re-evaluate the resident’s care plan.

Detecting Falls Using AI-Driven Video Monitors

Due to advances in technologies related to computer vision, there are many technologies that will detect falls.

A shout out to Dr. Larry Roberts whose doctorate thesis at MIT investigated the potential to extract three-dimensional information from 2D video images using mathematics back in 1960. His dissertation spawned numerous other research and development in video technology. In addition, the advancement in computer chips and artificial intelligence programming gave rise to what computer vision is today.

For the general public, this means having available video-based fall detection systems which is rapidly growing. It is an indication that these types of systems may become mainstream and replace fall-detection pendants and watches. However, there are distinct disadvantages to implementing computer vision technology as a fall prevention measure, as fall detection systems do not prevent falls:

Preventing Falls in the Future with Forecasting Tools

AI technology can also be used to prevent future falls through medication, diet, and exercise management or by providing real-time feedback to residents. Healthcare organizations can use population health forecast software that leverages AI to improve health outcomes. According to the American Health Association.

“Population health management refers to the process of improving clinical health outcomes of a defined group of individuals through improved care coordination and patient engagement supported by appropriate financial and care models.”

The benefits of population health software are well known and many healthcare providers including BlueCross BlueShield and others have made corporate investments in these solutions. For example, Zeomega, is supported by BlueCross BlueShield and Alliance Health. According to Definitive Healthcare, a provider of healthcare commercial intelligence, Epic Systems Corporation’s Healthy Planet PHM software is most widely used, followed by Oracle Cerner and IBM’s CareDiscovery Quality Measures and Explorys EPM Application Suite programs.

These are big corporate giants with a large footprint in the market. However, there are a number of smaller and more agile companies that are worth noting. For example:

Notable, the leading intelligent automation company for healthcare, launched Patient AI, the world's first use of large language models (LLMs), and GPT (the technology powering ChatGPT) to bring personalization at scale to healthcare.

Patient AI continuously reviews millions of data points across medical records and third-party data sets to develop a comprehensive clinical and social understanding of each patient. These insights are automatically translated into personalized recommendations that are surfaced to the patient before, after, and in between encounters through adaptive design without any staff involvement.

The result is that patients are engaged at precisely the right moment to take action in achieving their health goals.

‍Qventus uses AI, ML, and behavioral science to deliver the healthcare industry’s most powerful automation platform and solutions. According to Ian Christopher, co-founder, and chief technology officer, Qventus has

“ the ability to prioritize at-risk patients and bring the focus to those who need attention is rooted in AI technology".

The software, developed by Qventus, pulls data from electronic health records, then looks at nurse call-light and bed-alarm data, and finally combines it with other real-time information, such as medication and vitals, recorded by a nurse. When certain data elements line up, such as patients that set off the bed alarm and call light more than a certain threshold—which can vary depending on the ward, age, and medications given to a patient—it sets off a trigger. The system then sends out an alert to the nursing station, identifying that a particular patient is at a high risk of falling for the next 12 hours.

These companies use artificial intelligence to identify health risks and deliver forecasting tools for healthcare organizations to provide timely interventions so healthcare organizations can assist groups of people healthy as long as possible. These technologies help care organization to make deliver resources that empower their members with the means to take control and keep their health. They automatically identify risks, contact patients, notify their physicians, and recommend the next steps through changes in their diet, exercise, and/or medication.

One example of such innovation was introduced to my mom and dad who are in their mid-80s. Through the Medicare Advantage program and their physician, they were identified at a higher risk for hypertension and falls. The program provided them with a home health kit that included a scale, blood pressure cuff, and glucose monitor integrated into a mobile app. This enables their doctors to follow their health, remotely and automatically.

Benefits on Nursing Homes, Staff, and Health Outcomes

There are numerous benefits to using AI technology for fall detection and prevention in nursing homes. First and foremost, it can help keep residents safe and prevent injuries. Central to this blog post, falls can be particularly dangerous for older adults, and they can lead to long hospital stays and a decline in overall health. By detecting and preventing falls, AI technology can help improve health outcomes for nursing home residents.

In addition, AI technology can also help nursing home staff by reducing their workload and making their jobs easier. This can lead to increased job satisfaction and retention, as well as improved overall morale. Furthermore, it can help nursing homes save money by reducing the need for hospitalizations and other medical interventions.

One of the game-changing benefits of using AI technology for fall detection and prevention in nursing homes is that it can help address the healthcare worker shortages. With the aging population, there is an increased demand for healthcare workers in nursing homes. However, there is a shortage of healthcare workers, and this can make it difficult for nursing homes to provide the care their residents need.

To put things into perspective, the shortage of healthcare has long been anticipated since the late 1990s. Multiple organizations including the American Health Association, Center for Workforce Studies at the Association of American Medical Colleges, and others have reports and studies documenting this anticipated shortage. Even before the pandemic, people were leaving the healthcare workforce as shown by a report by Definitive Healthcare:

More recently, a US Health and News Report on July 2022 titled “Staff Shortages Choking U.S. Healthcare System” stressed, “a growing shortage of health care workers is being called the nation’s top patient safety concern.” The magnitude of this problem is evident and “the U.S. Bureau of Labor Statistics projects that more than 275,000 additional nurses are needed from 2020 to 2030, and that employment opportunities for nurses will grow at 9 percent, faster than all other occupations from 2016 through 2026.” To put things into greater perspective, the World Economic Forum predicts a shortage of 10 million healthcare workers globally by 2030, making this a global crisis.

By using AI technology for fall detection and prevention, nursing homes have the opportunity to reduce the workload of their staff. This, in turn, can make it easier for nursing homes to attract and retain staff, as they will be able to provide a safer and more manageable working environment.

What Are Experts Saying About the Viability of AI in Healthcare?

The opinions of domain experts believe that the viability of AI in health care is certain to impact how care is delivered and how we as consumers will benefit from them. For example, In 2020, the World Economic Forum Annual Meeting comprised of the foremost creative force for engaging the world's top leaders. In their annual meeting, the Chief Executive Officer, of Connected Care and Health Informatics, of Royal Philips reported the following:

"By 2030, AI will access multiple sources of data to reveal patterns in disease and aid treatment and care.
Healthcare systems will be able to predict an individual's risk of certain diseases and suggest preventative measures.
AI will help reduce waiting times for patients and improve efficiency in hospitals and health systems.”

In a 2021 report by KPMG, a global network of independent member firms offering audit, tax, and advisory services to mitigate risks and grasp opportunities in the industry addressed this very question.

According to their 2021 Thriving in an AI World report,

82% of healthcare executives say they would like their organization to move more aggressively in adopting AI.

Beyond the hype of AI brought to the full attention of the general public with access to chatgpt and opinions, an objective evaluation of AI adoption was presented in a September 2022 joint publication between the National Academy of Medicine and the U.S. Government Accountability Office.

Their report Technology Assessment, Artificial Intelligence in Health Care presents the benefits and challenges of machine learning technologies for medical diagnostics. The report highlights that despite the advancement of artificial intelligence to help medical professionals diagnose diseases resulting in benefits that include earlier detection of diseases; more consistent analysis of medical data; and increased access to care, particularly for underserved populations.''

AI was not widely adopted. Digging deeper into the challenge, they discovered that the barriers to adoption were (1) the need to demonstrate real-world performance, (2) meeting medical needs, and (3) addressing regulatory gaps, affecting technology developers, medical providers, and patients.Personally, I liken the growth of machine learning and artificial intelligence to the invention of the first gas-powered car invented by Carl Benz (yes, Mercedes Benz) in 1886. This fabulous invention would take about twenty years before the Ford Model T would be mass-produced and adopted. The barriers to adoption for the automotive industry was primarily the price and the public perception that it was for the wealthy. Similarly, the adoption of artificial intelligence to make it into mainstream healthcare and overcoming the barriers will take time.

Conclusion

In conclusion, AI technology has significant potential for fall detection and prevention in nursing homes. It can help keep residents safe, reduce the workload of staff, address healthcare worker shortages, and improve overall health outcomes. While there may be some challenges in implementing this technology, the benefits are clear, and nursing homes should consider investing in AI technology for fall detection and prevention.

Products that provide a real-time direct solution to prevent falls and fall-related injuries are lacking. There is a tremendous unmet need across all aspects of the healthcare continuum. Instead, they offer tools such as sensors, videos, and activity records to gather information and assess the probability of a patient falling in the far future. While caregivers can use this information to develop long-term fall prevention strategies, these products do not provide actionable steps to help seniors in real-time.

While the cause of falls may be multifactorial, the critical common theme is that they can't be prevented because there isn't a simple and cost-effective solution for caregivers and frontline healthcare staff to know if an older adult has imminent fall risk. This is a grave concern, as the repercussions of falls and fall injuries affect not only the individuals themselves but society and the economy of all nations, both small and large.

Learn more about how OK2StandUp Can help your nursing home improve patient safety.

If you are seeking a solution beyond the status quo. Beyond traditional approaches that require systems that detect falls or tools to create long-term fall prevention plans, I encourage you to learn more about OK2StandUP and how artificial intelligence is transforming the lives of how seniors can age safely wherever they choose to live. Moreover, how OK2StandUP can change the paradigm of care from reactive to proactive.

The Need For AI Technologies in Elder Care