Using technology to improve workflows; reduce clinical burnout

AI-powered sensing tech can assist care teams and prevent patient falls while improving nurse workloads and mitigate risks of patient injury.


Faced with a national shortage of nurses, healthcare organizations are hoping new technologies can help reduce nurses’ workloads, minimize staff burnout and lower costs.


monitoring patients for fall prevention represents a significant day-to-day nursing workflow burden and staffing expense.


For example, using artificial intelligence-powered sensing technology has the potential to prevent patient falls, and that can relieve nurses’ observation burdens while reducing the risk of patients getting injured.

In 2020, there were more 170,000 patient falls in U.S. hospitals, with more than 42,000 of those incidents resulting in mild or moderate harm, and more than 800 resulting in severe harm or death, according to the Agency for Healthcare Research and Quality. The Centers for Disease Control and Prevention has calculated that annual U.S. spending on non-fatal fall injuries totals $50 billion, while $754 million is spent on fatal falls.

Clearly, monitoring patients for fall prevention represents a significant day-to-day nursing workflow burden and staffing expense. 

Case study: Putting technology to use

A post-acute care organization in Wichita, Kan., is using the latest AI-powered sensing technology to help prevent resident falls, reduce expenses and mitigate adverse effects of nurse burnout.

Larksfield Place Retirement Communities, which offers independent living, assisted living and skilled nursing options, uses AI sensors to predict a potential patient fall 30 to 65 seconds in advance, giving care teams adequate time to intervene.

Step one: Quantify the patient fall risk

A critical first risk step in any effort to prevent patient falls is to conduct a patient fall risk assessment.

Compared with manual assessments, which are time-consuming for nursing staff as well as open to interpretation and human error, automated assessments are far more efficient; they can be completed within minutes. By having critical fall-risk information upon admitting a patient, care teams and therapists can devise tailored therapies and interventions that mitigate risks for falls.

To determine which residents are at high risk for falls, Larksfield Place’s staff works collaboratively with a local rehabilitation group. The team uses a fall risk assessment tool that incorporates AI and machine vision to objectively identify deficits in balance, gait and function – the three leading indicators of fall risk. The system automatically generates reports after each assessment so therapists can create patient-centered care plans.

Step two: Rely on intelligent surveillance

Residents in the facility’s short-term rehab unit who are identified as having a high risk of falling are further protected using AI-powered monitoring within the room.


AI-supported alerts result in less than one false alarm per day per bed, compared with the average of 15 false alarms per day with bed pads.


AI sensors detect a resident’s intent to exit a bed or chair and send alerts to the nursing team. This provides an extra layer of protection for residents at high risk of falling, relieving nurses from some of the monitoring burden. And this saves clinical time and cost.

The cost of telesitters (staff who video-monitor patients) to provide 24 hours of coverage for a floor can be $360 to $480 per day (based on a pay rate of $15 to $20 per hour). Using in-room sitters is far more costly. So, using sitters is financially unsustainable.

Although pressure pads can help provide alerts about patients’ movements, nurses have reported serious concerns because of false alarms (10 to 15 per day) and the limited response time that pads enable. Chasing and addressing false alarms is a top reason for nurse burnout and stress. The AI-powered sensors, in contrast, can more accurately detect as much as 65 seconds in advance that a patient intends to get up from a bed or chair. The system’s remote monitoring capabilities improve nursing efficiency and are particularly helpful during night shifts, when fewer staff members are in the building.

​The AI-supported alerts result in less than one false alarm per day per bed, compared with the average of 15 false alarms per day with bed pads. By detecting patients’ movements without the intrusiveness of a camera, the technology safeguards patient privacy.

Step three: Measure ROI

The AI-supported solution reduced falls resulting in injuries at Larksfield Place by 80 percent over the course of a study period. The organization also was able to significantly reduce the number of staff hours dedicated to monitoring patients.

The technology also can be extended for use into acute care settings, offering the promise of relieving those care teams of similar workload stresses. For example, one academic medical center using this technology recently reduced falls on a single nursing unit from 4.56 to 0.36 per 1,000 patient days. Falls with major injury decreased from just under one fall per month to zero. Finally, the organization’s monthly spend on fall prevention for the unit dropped by 66 percent.

The role of innovation

Emerging technologies, such as AI, can play an important role in monitoring patients and alerting care teams in advance of potentially harmful falls.

Many more AI-powered monitoring use cases, such as early warning systems for sepsis, are on the horizon. We must continue tapping new technologies and let innovation lead the way.


Karen Nelson, RN, is vice president of clinical services at Larksfield Place Retirement Communities in Wichita, Kan., a not-for-profit life plan community.

Tom Hale, MD, is the chief medical officer at VirtuSense Technologies, an AI-powered technology company.

More for you

Loading data for hdm_tax_topic #clinician-experience...