Cost Reduction

Using Artificial Intelligence to Reduce Long-Term Care Costs and Improve Patient Outcomes

October 3, 2017 9:09 am

In assisted living facility reduced transitions to nursing homes, resulting in $400,000 in savings.

The U.S. healthcare system is in crisis. Ten thousand Baby Boomers become eligible for Medicare every day, according to the Pew Research Center. This Silver Tsunami threatens to inundate a healthcare system that is already swamped with high costs and systemic inefficiencies. Not only is the current level of healthcare spending unsustainable (17.8 percent of GDP, as reported by the Centers for Medicare & Medicaid Services), but by 2050, the size of the over-65 population will nearly double, according to U.S. Census Bureau data. In addition, 92 percent of seniors will have at least one chronic condition and 77 percent will have two or more, according to the National Council on Aging.

The combined factors of aging with multiple chronic conditions is forcing us to rethink our healthcare delivery system. One potential solution is the ability to create value in our healthcare delivery by using technology to assist its least tapped resource system: patients and their at-home caregivers.

For the past two years, an innovative assisted living facility (ALF) in Lincoln County, Maine, has attempted to improve the care of its residents by equipping its caregivers with technology that tracks care provided, offers reminders to caregivers, and provides analysis on patients’ health statuses.

The technology contributes to improved care and fewer residents transitioning to higher-cost long-term care facilities: The net savings from aging in an ALF versus a nursing home is approximately $50,000 per year.

Challenges Posed by an Aging Population

Maine has the oldest population of any state in the country with a median age of 44.6 years, and as one of the oldest counties in the state, Lincoln County is especially threatened by rising healthcare demands, according to the Census Bureau.

With a growing aging population that is living longer and a declining workforce, additional burden will be placed on caregivers to support the 70 percent of Lincoln County’s senior population who are expected to need long-term services and support, according to data from the Office of Aging and Disability Services, Maine Department of Health and Human Services.

To address its long-term services and support needs, Lincoln County relies on private non-medical institutes and adult family care homes such as ALFs. ALFs provide housing and support services to residents who can no longer live independently but do not require specialized medical services that would be provided in a nursing home.

Seven ALFs (one facility served as the control without the assistive AI technology)in Lincoln County participated in a recent study. Each is an adult group home with six to eight residents in each home. Non-clinical personal care attendants assist with activities of daily life such as meals, medication, and bathing. Residents receive emotional support and remain socially engaged in a setting that they consider “the next best thing to being home.”

Technology as an Assistive Solution

The ALF sought to use technology to address the following operational needs:

  • Easier caregiving with fewer errors
  • A fully digital record of the care that is provided
  • Fewer patient days away at nursing homes or hospital.

The chosen technology has the following attributes.

Reminders for caregivers. Caregivers are reminded when to provide care and what care is needed for each resident, including medications, vitals monitoring, symptoms checking, and assistance with activities of daily living, including meals and exercise. The reminders are customized for each resident and driven by a care plan that is specified by the resident’s primary care physician. These reminders relieve the caregivers from the burden of remembering the complex sequence of tasks they need to deliver for each resident.

Tracking. The system digitally tracks all care activities. A supervisor can quickly assess and address gaps in caregiving.

All data in the system is graphed in real time so it is easy to understand and provides staff with visual interpretation of vital trends. Caregivers can quantitatively see which residents are at risk or experiencing difficulties and direct attention accordingly.

Analysis. The system analyzes all vital signs and symptoms in real time to recognize when health issues need clinical intervention. This empowers and enables non-clinical personal care attendants to reach out for clinical support from the resident’s primary care physician in a timely fashion.

With limited medical backgrounds and several residents to care for, caregivers at the ALF cannot be expected to know the details of every individual’s care plan. With the proper technology, caregivers develop a deeper understanding of the needs of their residents, helping them deliver timely, precise, and personalized care.

For example, one caregiver reported that she likes to be alerted about health exacerbations and stated that the chronic obstructive pulmonary disease (COPD) protocol “proved itself very helpful and effective for an emergency situation. The alerts helped me to monitor a resident and her symptoms before clinical staff were even aware of a COPD flare up.”

Study Demographics

Over a two-year period, the technology was used to care for 70 ALF residents. Most residents are more than 80 years old and have one or more chronic conditions with the following distributions: 14 with depression, eight with COPD, 12 with congestive heart failure, 15 with diabetes, one with autism, and 24 with hypertension.

Significant Outcomes

Since the program start, technology has assisted ALF caregivers for more than 23,000 resident days. During that time, caregivers were reminded to perform 475,115 tasks. Thorough care saves costs because conditions are managed or never materialize.

As a result of the reminders, a significant number of vital signs were recorded and symptoms were checked, resulting in 251 health exacerbations detected. Of the 251 health exacerbations, 59 were from COPD flare-ups, 80 were hypertension episodes, 72 hyperglycemic events, 12 hypoglycemic events, and 18 fluid retention situations. The ALF addressed the exacerbations in a timely fashion by following step-by-step instructions provided by the technology with clinical oversight from residents’ primary care physicians if needed.

The net benefit of rigorous compliance with individual care plans and early interventions in the face of patient health declines is clearly evidenced when we analyze residents’ end-of-life circumstances based on whether they were cared for with the technology (see the exhibit at the bottom of page 7).

During the two-year study, 31 residents left the ALF. We compared them to 34 ALF residents with similar demographics who left during the two years prior to the study. The contrast is significant. Prior to the technology introduction, 22 residents left the ALF permanently to move to a nursing home or hospital. In contrast, when supported by technology, only 12 were moved to a nursing home and none left to be hospitalized. That is a 45 percent increase in aging in place. The average and median ages were 2 or more years older for those who received technologically assisted care.

The net savings from aging in an ALF versus a nursing home is approximately $50,000 per year. Average duration of a nursing home stay is two years. Therefore, the total savings from four fewer transitions to the nursing home is $400,000. There are also significant savings associated with six fewer hospital end-of-life transitions.

The Power of Empowering Caregivers

As evidenced by this study, the ability to age at home increases when caregivers are aided by technology. Caregiving is relentless, exhausting, and stressful, and it gets harder over time as patient care becomes more complicated. This study suggests that when caregiving is made easier through intelligent technology, life is extended and end of life is more likely to occur at home rather than in an institutional setting.


googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text1' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text2' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text3' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text4' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text5' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text6' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-text7' ); } );
googletag.cmd.push( function () { googletag.display( 'hfma-gpt-leaderboard' ); } );