Avoiding Alert Fatigue and Clinician Burnout While Scaling Virtual Care

By Simon MacGibbon

We know a virtual care system for patients and clinicians is immensely beneficial, empowering patients to take control of their health, supporting optimal clinical management, and predicting and preventing life-threatening events through reliable intervention. Mercy Health’s virtual hospital Mercy Virtual has reduced hospitalizations by approximately 50%, for instance, and has significantly lowered the overall cost of care for high-risk patients living with chronic diseases. 

The ability to capture and surface data through remote monitoring is a necessary component of a high impact virtual care program. If used effectively, virtual care can dramatically improve patient outcomes, but at what cost? In this article, we discuss a critical obstacle to scaling virtual care that has the potential to dampen its transformative expectations. 

Addressing Alert Fatigue and Clinician Burnout

Physicians experience burnout at alarming rates and more than any other category of U.S. workers, according to recent studies reported in the Kansas Journal of Medicine. As of 2019, 44% of physicians reported suffering from burnout, 11% felt colloquially depressed, and 4% were clinically depressed from some job-related factor.

Burnout – and the rise of alert fatigue – is one of the biggest complications in scaling virtual care. Alert fatigue, however, isn’t a new concept. In fact, a recent survey with input from over 15,000 physicians, showed that the top contributors to burnout have been linked to administrative/clerical tasks, hours of work, and the electronic health record (EHR).

In hospital units, a Journal of Medical Internet Research paper found that 80-99% of alarms are false or clinically insignificant and do not represent real danger for patients, causing clinicians to miss relevant alarms and experience ongoing, unnecessary alert fatigue. With the implication of high quality Machine Learning, we can curve the false positive rate and actually recognize urgent alerts in order to service the patients in critical need. 

With the inevitable rise of virtual care – with patient data flowing freely from the home – alert fatigue is a looming complication for the healthcare system. Remote Patient Monitoring (RPM) reimbursement codes do not account for the operational realities of scaling virtual care.

The associated clinical cost to serve, the scarcity of clinical resources, and potential clinician burnout are all problems facing the virtual care revolution, but with the right AI and Machine Learning technology, we can actually prevent these issues and ensure that patients living with chronic disease get access to continuous, preventive care without overwhelming overworked and low-staffed clinical offices.  

Implementing Five Elements for Scalability

While virtual care is the future, scaling it will require acceptance from health care providers, many of whom are resistant to new technologies, according to studies, fearing that the patient-clinician relationship will suffer, they will no longer have management control, and diagnoses will be disrupted. Whether they’ve accepted this yet or not, AI and Machine Learning will actually support their relationships, provide them with more control, and allow for more reliable patient care. 

Scaling virtual care will require a combination of technology, human capital, and clinical guidelines. By following these steps, we can achieve this, while lowering burnout rates, producing better patient outcomes, and achieving more beneficial cost savings:

  1. Recognize individualized baselines. Using a sound statistical approach to computing individual physiological and functional baselines, by monitoring entropy in patient data such as vital signs, will allow computational assessments to recognize – and flag – their significance. For instance, one heart failure patient might experience frequent changes in weight that revert without clinical consequence, making weight gain less concerning than a heart failure patient who suddenly experiences an abnormally high rate of weight gain. Systems should be built to account for these individual differences. 
     
  2. Establish clear safety thresholds. This will effectively guarantee a fail-safe design together with the individualized baselines. It will also be used to incorporate established medical guidelines to prevent any one person from dismissing the system. Fail-safe designs are an important feature of automated systems and virtual care should not be an exception. 
     
  3. Build a learning system that has the facility to be overridden by clinicians. Already, clinicians use non-medical sticky notes and spreadsheets to take note of potential problems which aren’t often caught by medical monitoring (ex. patient looks more pale than normal; worried patient may not tolerate this new medication). The right virtual care system takes this into account, making it easy for clinicians to apply their own judgement.  
     
  4. Use machine learning to make sense of a wide range of data and its aggregate meaning. While medical professionals have been trained to provide expert analysis on a patient’s condition based on a combination of factors, the human brain is not designed to make sense of high dimensional datasets. This is the ideal application for Machine Learning, to augment clinicians to see concerning changes in patient health status that might otherwise go unnoticed.
     
  5. Take a data driven individualized approach to establishing scheduled visits. Just like the in-patient setting, unplanned alerts ought not be the only way patients are surfaced. This is critical for patients with chronic disease. While patient data should be used to monitor a patient’s condition, periodic visits (virtual, if possible) are essential for those living with chronic conditions. AI doesn’t detract from that, but rather supports the ongoing patient-clinician relationship. 
     

The virtual care model should maintain a digital profile for each patient, which takes normal patterns of volatility into account, while analyzing numerous factors (vitals, labs, demographics, history, etc.) to make a clinician’s job in monitoring, treating, and following up with patients less time consuming and more valuable, ultimately creating a better system. Facilitating the transition to making virtual care the standard model isn’t going to happen overnight, but we anticipate that it will, in time, transform our current understanding of health care.

In essence, virtual care technology companies need to take ownership of challenges like alert fatigue and make systems that are operationally scalable. To join the revolution, we encourage health care providers to begin the process of acceptance and start to build a team of clinicians who are ready to embrace this new, forward-thinking model.  

To learn about data driven virtual care, visit Myia Health. To see how Myia Health is building the operating system for data driven virtual care, please click here.

No comment yet, add your voice below!


Add a Comment

Your email address will not be published. Required fields are marked *