Peoria, Illinois,
16:17 PM

Common medical statistics often wrong or misleading

Groundbreaking solution co-authored by OSF HealthCare Senior Fellow holds great promise for improving AI in patient care

Image for predictive analytics

A simple, yet revolutionary new statistical technique enables better assessment and implementation of many tests and predictive models, leading to greater patient benefits. Faulty assumptions in some widely used statistics can lead to flawed predictive model implementations that impact patient care. This can be corrected with a novel, utility-based approach (“u-metrics”), according to a study in the IEEE Journal of Biomedical and Health Informatics, co-authored by Dr. Jonathan Handler, senior fellow for Innovation at OSF Healthcare.

What’s wrong with the classic statistics?

Predictors of whether something will or will not happen in the future are used to facilitate care. Classic statistics to assess these predictors and guide their implementations include sensitivity, specificity, and positive and negative predictive values. These are based only on counts of how often the predictor was right or wrong. The article notes that these classic statistics make assumptions that don’t apply to many (probably most) real-world scenarios. Statistics based on faulty assumptions might suggest that a predictor will yield great benefit to patients even though the real-world performance will prove disappointing or even harmful. The result? Too often, busy health care workers must suffer through frequent false or useless alarms that they soon learn to ignore (“alert fatigue”).

For example, a prediction system might incorrectly trigger an alarm, claiming that a patient who is healthy has a dangerous infection. It might also correctly trigger an alarm for a patient with a dangerous infection even though the team is already addressing the issue. In each case, the alarm adds no value and distracts the care team away from other important work. Worse, in the case of a correct but unhelpful and distracting alarm, classic statistics inappropriately “take credit” for a correct prediction even though the alarm created more harm than benefit. This is because classic statistics assume that correct predictions are always helpful and every correct prediction is equally helpful, even though, as the authors note, those assumptions are commonly not the case.

A new and better approach

To address these challenges, the authors created u-metrics, an intuitive and comprehensive solution that does not rely on assumptions that rarely apply in the real world. Unlike classic statistics, it does not take a one-size-fits-all approach. Instead, it assigns to each prediction only the credit it deserves, and categorizes each prediction based on the benefit or harm created rather than its correctness.


Dr, Jonathan Handler, Senior Fellow, OSF Innovation

“Health care providers often complain that research suggests predictors will perform well, but when implemented in the real world, the impact is disappointing and sometimes harmful.” 

Dr, Jonathan Handler, Senior Fellow, OSF Innovation

“There has been some limited acknowledgement that the assumptions of classic count-based statistics commonly do not apply and a few partial fixes have been proposed. However, to our knowledge, this is the first comprehensive evaluation of the assumptions required by these classic statistics, and more importantly, the first comprehensive fix to the problem.” Dr. Handler added, “We believe that using u-metrics to guide the development, selection, and implementation of these types of predictors will benefit patients, providers, and health systems.”

In addition to clinical predictors, u-metrics can be used to assess any system that provides yes or no responses, from weather alerts to stock market predictions.

How can u-metrics help address alert fatigue?

By better informing the selection of predictors and their implementations, u-metrics might reduce the likelihood that a health system will choose and operationalize a predictor that fires too many useless alerts. The paper also describes “snoozing,” an implementation technique to dramatically reduce false and nuisance alerts in many cases. Snoozing is when an alarm is automatically or manually silenced for a period of time after it fires. Although manual snoozing of sensor alarms is common in ICUs, its use has not been well studied for predictive alarms. The paper notes this might be due to the inability of classic metrics to correctly assess the impact of snoozing. The u-metrics solution correctly assesses the impact of snoozing because it will neither reward nor penalize the system for suppressing alerts that were correct but would create distraction or harm if fired. The paper also describes a method to identify optimal snooze times. Properly applied, snoozing could reduce alert fatigue and increase the likelihood that clinicians will respond to true alarms.

Looking for more?

Dr. Handler provides an accessible explanation of u-metrics in his blog: Shocking new discovery: recall and sensitivity are not the same! as well as an explanation of the techniques to reduce false positives in his blog: Avoiding false alerts: snoozing ≠ laziness.

Healthleaders Media recently hosted a panel discussion about artificial intelligence featuring Dr. Handler. Here’s a link to the webinar. Dr. Handler discusses the importance of using better metrics when developing predictive models at (36:34) minutes into the webinar.

Dr. Handler leads OSF HealthCare’s Clinical Intelligence lab, with a mission to help patients get correct diagnoses and optimal therapy as quickly as possible. Visit to learn more about how OSF HealthCare is working to transform health care.

Video clips with Dr. Jonathan Handler, senior fellow  for Innovation at OSF HealthCare (from Healthleaders Media webinar on Medical Use of Artificial Intelligence)

View Dr. Jonathan Handler-Models often overpromise and underperform
Dr. Jonathan Handler-Models often overpromise and underperform
View Dr. Jonathan Handler-Models can drive annoying alertsmp4
Dr. Jonathan Handler-Models can drive annoying alertsmp4
View Dr. Jonathan Handler-New metrics improve models.mp4
Dr. Jonathan Handler-New metrics improve models.mp4

OSF HealthCare is an integrated health system owned and operated by The Sisters of the Third Order of St. Francis, headquartered in Peoria, Illinois. OSF HealthCare employs nearly 24,000 Mission Partners in 150 locations, including 15 hospitals – 10 acute care, five critical access – with 2,089 licensed beds, and two colleges of nursing throughout Illinois and Michigan. The OSF HealthCare physician network employs more than 1,500 primary care, specialist and advanced practice providers. OSF HealthCare, through OSF Home Care Services, operates an extensive network of home health and hospice services. It also owns Pointcore, Inc., comprised of health care-related businesses; OSF HealthCare Foundation, the philanthropic arm for the organization; and OSF Ventures, which provides investment capital for promising health care innovation startups. More at